Simulated Robots Package

Simulation for differential drive robots using ROS2 Jazzy and Gazebo Harmonic. This package provides all of the necessary files to get a simulated robot up and running. This includes the urdf, parameters and launch files for a robot with a lidar sensor and tele-operated navigation. More sensors and functionalities will be added in future.

Branches

This repository has two main branches:

• main: Base repository with core robot simulation functionality including URDF, Gazebo simulation, and basic tele-operation
• mapping: Extended branch that adds mapping capabilities using SLAM Toolbox and robot localization with Extended Kalman Filter (EKF)

Work in progress

The package is still being worked on and in development

Supported on

Supported for Ubuntu 24.04 & ROS2 Jazzy but compatibility with other versions has not been checked.

Install Required ROS 2 Packages

Base Packages (Required for all branches)

Make sure to install the following ROS 2 Jazzy Packages:

sudo apt install -y                         \
   ros-jazzy-ros-gz                        \
   ros-jazzy-ros-gz-bridge                 \
   ros-jazzy-joint-state-publisher         \
   ros-jazzy-xacro                         \
   ros-jazzy-teleop-twist-keyboard         \
   ros-jazzy-teleop-twist-joy 

Additional Packages (Required for mapping branch)

If you're using the mapping branch, also install:

sudo apt install -y                         \
   ros-jazzy-slam-toolbox                  \
   ros-jazzy-robot-localization

Install

To use this package please download all of the necessary dependencies first and then follow these steps:

Main Branch

For the base robot simulation, clone the main branch:

mkdir -p ros2_ws/src
cd ros2_ws/src
git clone https://github.com/adoodevv/diff_drive_robot.git
cd ..
colcon build --packages-select diff_drive_robot --symlink-install

Mapping Branch

To use the mapping features, clone the mapping branch directly:

mkdir -p ros2_ws/src
cd ros2_ws/src
git clone -b mapping https://github.com/adoodevv/diff_drive_robot.git
cd ..
colcon build --packages-select diff_drive_robot --symlink-install

Usage

Main Branch - Basic Robot Simulation

After sourcing ROS and this package, launch the 2-wheeled differential drive robot simulation:

source install/setup.bash
ros2 launch diff_drive_robot robot.launch.py 

Controlling the robot

Currently, only keyboard control works. Run this in another terminal:

ros2 run teleop_twist_keyboard teleop_twist_keyboard 

Mapping Branch - Mapping and Localization

The mapping branch extends the base functionality with:

• SLAM Toolbox: For mapping the environment
• Extended Kalman Filter (EKF): For sensor fusion and improved odometry estimation
• Enhanced RViz configuration: Pre-configured for mapping visualization

New Configuration Files

The mapping branch includes additional configuration files:

• config/slam_toolbox_mapping.yaml: Configuration for SLAM Toolbox mapping mode
• config/ekf.yaml: Extended Kalman Filter parameters for sensor fusion (odometry and IMU)

Launching the Mapping System

The mapping system requires launching two separate terminals:

Terminal 1 - Robot Simulation:

source install/setup.bash
ros2 launch diff_drive_robot robot.launch.py

This launches:

• Gazebo simulation
• Robot State Publisher
• Gazebo-ROS bridge
• Extended Kalman Filter node
• Robot spawner

Terminal 2 - Mapping:

source install/setup.bash
ros2 launch diff_drive_robot mapping.launch.py

This launches:

• SLAM Toolbox (online async mapping mode)
• RViz with mapping configuration

Controlling the robot during mapping

In a third terminal, run the keyboard teleop:

ros2 run teleop_twist_keyboard teleop_twist_keyboard

Saving the Map

Once you've mapped your environment, you can save it using the SLAM Toolbox plugin in RViz or via command line:

ros2 run nav2_map_server map_saver_cli -f ~/my_map

TODO

Package is still being worked on, though the core functionality is pretty much done, I will be adding some more sensors and functionalities soon.